apkk download link:https://www.upload-apk.com/jXAIGRCMOVLCrNM

Telemedicine Compose Prototype

A Jetpack Compose single-activity Android prototype demonstrating a multilingual telemedicine experience for three roles: Patient, Doctor, and Jan Aushadi Worker. The UI is wired end-to-end with mock data, voice-to-text, and a pluggable symptom prediction client so the app can be demoed without any backend services.

Highlights

• Role-based login: pick Patient, Doctor, or Jan Aushadi Worker to explore contextual dashboards.
• Global language toggle: every screen switches between English and Hindi instantly, including top app-bar controls.
• Patient tools: health record snapshots, text/voice symptom checker with mock AI predictions, pharmacy stock, and the shared video-call surface.
• Doctor workflow: daily schedule, live call queue with accept/reject state, and quick dial room that launches the dummy video UI.
• Jan Aushadi worker tools: manage medicine orders, progress in-memory status, and edit inventory counts.
• Voice first symptom flow: Hindi speech → transcription → translation → mock API request, with graceful error handling and permission prompts.
• Development screenshot loader: optional card that reads customer reference screenshots from external storage during demos.

Run Instructions

1. Open the project in Android Studio Ladybug or newer.
2. Let Gradle sync and run app on an emulator or Android device (API 24+).
3. Grant microphone and photo-library permissions when prompted. Voice input and screenshot previews depend on them.

Matching customer screenshots during a demo

1. Copy the saved UI screenshots from the development machine to the emulator/device path:

   adb push "C:\Users\Kathansh Jain\Saved Games\*" /sdcard/Pictures/SavedGames/

2. In the Patient overview screen tap Load demo screenshots; the app will display any PNG/JPG/WEBP files it finds in /sdcard/Pictures/SavedGames/.

Configuring API keys (optional)

Mock implementations ship enabled by default. To swap in real services, add keys to local.properties or pass them as Gradle properties:

SYMPTOM_PREDICTOR_API_KEY=your-live-key
TRANSLATION_API_KEY=your-translate-key

• BuildConfig.SYMPTOM_PREDICTOR_BASE_URL currently targets https://api.symptom-predictor.example/. Update app/build.gradle if your backend uses a different hostname.
• The SymptomPredictorRepository uses an OkHttp interceptor that returns canned data. Remove the interceptor and provide a real Retrofit implementation when integrating.
• Translation currently relies on ML Kit on-device models; if you wire a REST service instead, read the key via BuildConfig.TRANSLATION_API_KEY inside TranslatorManager.

Permissions requested

• RECORD_AUDIO – required for Hindi voice capture in the symptom checker.
• READ_MEDIA_IMAGES / READ_EXTERNAL_STORAGE – debug-only feature for loading reference screenshots.
• INTERNET – needed for Retrofit/OkHttp even though responses are mocked locally.

Project structure (high level)

• MainActivity hosts the navigation graph and global language state.
• ui/screens contains Compose screens per role.
• network hosts the Retrofit models, service interface, and mock repository.
• voice and translations wrap the SpeechRecognizer and ML Kit translator.
• data/MockData.kt contains immutable sample records used across the app.

Testing notes

This is a UI prototype without instrumentation tests. Manual walkthroughs are recommended after any changes:

• Launch each role from the login screen and step through all destinations.
• Toggle English ↔ Hindi from every top-bar.
• Run both text and voice symptom checks (voice flow requires a microphone).
• Confirm doctor call requests reflect Accept/Reject state and open the mock video UI.
• Load development screenshots to verify permission copy/paste instructions.

Enjoy the demo! The code is purposely annotated and modular so you can replace the mock layers with real services when the backend is ready.